News for hackers. Stuff that matters.

goc

Re:Jeff Dean on Large-Scale Deep Learning at Google on Wednesday March 16, 2016 @01:48PM (#11299245)
I am very interested in AI that can teach itself(sounds too great). Where can I learn up about such AI(related concepts and the whole 9 yards) to start reading papers in the field? I am just looking for comprehensive sources(preferably textbooks).
Questions in Epp are by no means unique. If you search MSE, you'll see that every question in Epp has probably been asked and reasked about a thousand times each. That goes for subjects like Real Analysis, Abstract Algebra, Topology as well.
To start from absolute zero, check out Suzanna Epp's Discrete Math[0]. I believe even a motivated high school student could get started with it and even finish it. If your proof-writing is shaky, the book provides a very good workout. From there it will be easy to choose the areas of discrete math to specialize.

[0] http://www.amazon.com/gp/product/0495391328?keywords=susanna...

Any one know any books or resources that concentrate on invariant programming? Translating recursive code into properly tail recursive or iterative code can be pretty difficult.
Re:Ask HN: Book Recommendations? on Saturday March 05, 2016 @08:14PM (#11231990)
In this video[0] Van Roy claims that the function Three "resolves" to 3. It's at around 8:45 minute. I tried working that out, but can't seem to get it correct. What I tried

   {Inc {Inc {Inc Zero}}}
   {Inc {Inc 1}}
   {Inc {1} + 1}
   {{{1} + 1} + 1}
Can we consider {{{1} + 1} + 1} to be 3 or what are my mistakes? Thanks.

[0] https://youtu.be/Arh_6J5_fWU?t=8m46s

Re:Ask HN: Book Recommendations? on Saturday March 05, 2016 @09:43AM (#11229614)
I am reading CTM by Van Roy/Haridi and watching Van Roy's edx vids. Very nice explanations that require a lot of filling in the holes.

If someone else is familiar with Oz/Van Roy stuff, I have a question.